Seismic Resistance of Square Concrete Columns Retrofitted with Glass Fiber-Reinforced Polymer

ACI Structural Journal, Sep/Oct 2005 by Memon, Muhammad S, Sheikh, Shamim A

An overview of the two sets of columns discussed above shows that the higher ductility, higher energy dissipation capacities, and enhanced shear and moment capacities were obtained through GFRP strengthening of deficient columns. Regardless of the ductility or toughness parameters considered, there was a positive relationship between improved column performance and increasing GFRP retrofit layers.

Effect of GFRP retrofitting on damaged columns

To evaluate the performance of damaged columns repaired with GFRP, the responses of Specimens ASG-2NSS and ASGR-7NSS are compared (Fig. 9 and 14). Specimen ASG2NSS was wrapped with two layers of GFRP and tested under an axial load of 0.33P^sub o^, whereas Specimen ASGR-7NSS was first damaged before being retrofitted with two layers of GFRP and then tested to failure under the same level of axial load. The curvature ductility factor and cumulative curvature ductility ratios (Table 5) for Specimen ASG-2NSS are 20 and 45% higher than those for Specimen ASGR-7NSS. The energy and work damage indicator values for Specimen ASG-2NSS are 76 and 21% higher than those of the previously damaged Specimen ASGR-7NSS. The significant reduction in the total ductility parameters reflects the previous damage sustained by Specimen AS-7NS prior to retrofit. Nevertheless, considering the overall performance, the behavior of the repaired Specimen ASGR-7NSS was significantly better than that of a similar unretrofitted Specimen AS-1NS (Fig. 16).

The comparison between the responses of Specimens ASG-6NSS and ASGR-8NSS (Fig. 13 and 15) is made to evaluate the performance of a damaged column retrofitted with GFRP under higher levels of axial load (0.56Po). Specimen ASG-6NSS was strengthened with six layers of GFRP before testing whereas Specimen ASGR-8NSS was first damaged then retrofitted with six layers of GFRP and finally tested to failure. Specimen ASG-6NSS sustained 14 cycles of lateral excursions compared to 11 cycles for Specimen ASGR-8NNS. The curvature ductility factor of the repaired Specimen ASGR-8NSS was 30% less than that of the strengthened Specimen ASG-6NSS. The toughness demonstrated by Specimen ASGR-8NSS, as indicated by the energy and work damage indicators, was between 17 and 30% of that of Specimen ASG-6NSS. The difference in performance between the repaired and the strengthened columns is more pronounced under high axial load (0.56P^sub o^) than under low axial load (0.33P^sub o^). The extent of damage prior to repair plays an important role in determining the column performance. A severe prerepair damage resulted in poor performance of the repaired column. The performance of repaired Column ASGR-8NSS, despite its extensive prerepair damage, is almost similar to that of Specimen ASG3NSS that was strengthened with four layers of GFRP.

GFRP jacketing improves the seismic behavior of previously damaged square columns although the amount of damage sustained previously greatly affects their repair potential and salvageability. The data also indicate that a more heavily damaged column requires a higher amount of GFRP to perform in a manner similar to that of an undamaged column. The repaired specimens were seismically superior to their control unwrapped counterparts. The GFRP retrofitting techniques would be particularly useful for restoring columns that have suffered light to moderate damage during an earthquake.